Thin film photovoltaic (PV) modules (also referred to as “solar panels”) based on cadmium telluride (CdTe) paired with cadmium sulfide (CdS) as the photo-reactive components are gaining wide acceptance and interest in the industry. CdTe is a semiconductor material having characteristics particularly suited for conversion of solar energy to electricity. For example, CdTe has an energy bandgap of about 1.45 eV, which enables it to convert more energy from the solar spectrum as compared to lower bandgap semiconductor materials historically used in solar cell applications (e.g., about 1.1 eV for silicon). Also, CdTe converts radiation energy in lower or diffuse light conditions as compared to the lower bandgap materials and, thus, has a longer effective conversion time over the course of a day or in cloudy conditions as compared to other conventional materials. The junction of the n-type layer and the p-type layer is generally responsible for the generation of electric potential and electric current when the CdTe PV module is exposed to light energy, such as sunlight. Specifically, the cadmium telluride (CdTe) layer and the cadmium sulfide (CdS) form a p-n heterojunction, where the CdTe layer acts as a p-type layer (i.e., a positive, electron accepting layer) and the CdS layer acts as a n-type layer (i.e., a negative, electron donating layer).
A transparent conductive oxide (“TCO”) layer is commonly used between the window glass and the junction forming layers. For example, the TCO layer may be sputtered from a cadmium stannate (i.e., Cd2SnO4) target by either of two processes: hot sputtering or cold sputtering. When hot sputtered, the TCO layer is typically deposited at sputtering temperatures above about 250° C. in a one step sputtering process. When cold sputtered (e.g., at about room temperature), the TCO layer must be annealed following sputtering of the layer in a second step to convert the layer from an amorphous layer to a crystalline layer.
Though the hot sputtering process is more streamlined (i.e., only requiring a single step), the hot sputtered TCO layers can have a much higher resistivity than the cold sputtered TCO layers—even when sputtered from the same material (e.g., cadmium stannate)—making the hot sputtered TCO layer less attractive for the end use. Although not wishing to be bound by any particular theory, it is believed that this difference in resistivities between the hot sputtered layer and the cold sputtered layer likely stems from a difference in the as-deposited stoichiometry. For example, when sputtering from a cadmium stannate target, it is presently believed that cold sputtering produces a layer having the stoichiometry Cd2SnO4, which is the desired stoichiometry for cadmium stannate. However, other processing issues exist that hinders the viability of cold sputtering to form the TCO layer, especially from a cadmium stannate target. First, the annealing process can sublimate cadmium atoms off of the TCO layer, altering the stoichiometry of the TCO layer, especially along its outer surface. Second, the annealing process can cause cracks and/or delamination of the TCO layer on the substrate, probably forming while increasing the substrate temperature from the sputtering temperature to the anneal temperature, while cooling from the anneal temperature back to room temperature, and/or from the phase change from amorphous to crystalline which also may result in a density/volume change. These two processing issues can lead to the stability problems in the TCO layers of the resulting PV devices, whereas such issues are not present in PV devices formed with TCO layers formed via hot sputtering.
On the other hand, hot sputtering from a cadmium stannate target is presently believed to produce a layer having the stoichiometry CdSnO3+SnO2. Thus, the TCO layer includes a deficiency of cadmium, when compared to stoichiometric cadmium stannate. This change in the as-deposited stoichiometry is believed to be a result of cadmium atoms sublimating off of the surface of the substrate during the hot sputtering process.
Thus, a need exists for a TCO layer having the conductivity of those cold sputtered layers with the processing and device stability found in those hot sputtered layers.